U.S. patent number 4,763,977 [Application Number 06/690,099] was granted by the patent office on 1988-08-16 for optical fiber coupler with tunable coupling ratio and method of making.
This patent grant is currently assigned to Canadian Patents and Development Limited-Societe. Invention is credited to Kenneth O. Hill, Derwyn C. Johnson, Masao Kawachi, Brian S. Kawasaki.
United States Patent |
4,763,977 |
Kawasaki , et al. |
August 16, 1988 |
Optical fiber coupler with tunable coupling ratio and method of
making
Abstract
An optical coupler for single mode optical signals having a
tunable (variable) coupling ratio, and a method of fabricating the
coupler. A pair of virtually identical optical fibers are fused
together at a narrowed waist region, each fiber being formed of a
core and cladding, each being tapered toward the waist in such a
manner as to ensure adiabatic propagation of light in the
structure. An optical signal carried by one fiber first passes
through a decreasing taper region and then passes through an
increasing taper region of one or the other fiber (or both). The
decreasing taper rate of the input portion of one fiber is such
that the optical signal radiates out of the core (where V=1
locally) and into the cladding, as it approaches the waist. The
increasing taper rates of the output portions of the two fibers are
such that a predetermined coupling ratio is obtained. The coupler
is bent in the region of the waist whereby a coupling ratio can be
selected between the incoming fiber portions having the decreasing
taper and the outgoing fibers having increasing tapers.
Inventors: |
Kawasaki; Brian S. (Kanata,
CA), Kawachi; Masao (Mito, CA), Hill;
Kenneth O. (Kanata, CA), Johnson; Derwyn C.
(Nepean, CA) |
Assignee: |
Canadian Patents and Development
Limited-Societe (Ottawa, CA)
|
Family
ID: |
24771080 |
Appl.
No.: |
06/690,099 |
Filed: |
January 9, 1985 |
Current U.S.
Class: |
385/43; 359/900;
385/42; 385/50; 385/51 |
Current CPC
Class: |
G02B
6/283 (20130101); G02B 6/2835 (20130101); Y10S
359/90 (20130101) |
Current International
Class: |
G02B
6/28 (20060101); G02B 006/26 (); G02B 006/42 () |
Field of
Search: |
;350/96.15 |
References Cited
[Referenced By]
U.S. Patent Documents
Primary Examiner: Sikes; William L.
Assistant Examiner: Wise; Robert E.
Attorney, Agent or Firm: Pascal & Associates
Claims
We claim:
1. A unidirectional optical fiber coupler comprising a pair of
virtually identical optical fibers each comprised of a core and
cladding and being fused together at a narrow tapered waist, each
fiber being tapered toward and away from the waist, means for
translating an optical signal in single mode and carried by one
fiber in the core at one side of the waist to multi-mode carried by
the cladding at least adjacent the narrowest portion of the waist,
and for translating the signal in multi-mode to single mode carried
by the core of either said one or the other fiber or both fibers at
the other side of the waist.
2. An optical fiber coupler as defined in claim 1 in which the
coupler is bent in the region of the waist.
3. An optical fiber coupler as defined in claim 2 further including
means for varying the angle of bending of said coupler whereby the
optical signal is selectively switched whereby it is carried by
either one of said one or the other of said fibers on the other
side of the waist.
4. An optical fiber coupler as defined in claim 3 including means
for restraining the fibers in either one of a U or S curve in which
the waist is within the curve, and means for varying the radius of
the curve, whereby the optical signal is switched and is
substantially coupled from said one to the other of the fibers.
5. An optical fiber coupler as defined in claim 3 in which the
rates of taper of the fibers in the region of said waist are
adiabatic, the minimum diameters and the refractive indices of the
cores and cladding in the region of the waist portion being such
that: V.ltoreq.1 where ##EQU3## in which V is the normalized
frequency (local) of the coupler (dimensionless),
a is the (local) radius of the core,
.lambda. is the wavelength of an optical signal passing through the
coupler and
n.sub.core and n.sub.cladding are the indices of refraction of the
core and cladding respectively,
but that V associated within the waveguide formed by the
air-cladding interface is greater than a value in which the
antisymmetric transmission mode of the cladding waveguide radiates
out of the cladding.
6. An optical fiber coupler as defined in claim 2 including means
for restraining the fibers in either one of a U or S curve in which
the waist is within the curve, and means for varying the radius of
the curve, whereby the optical signal is switched and is
substantially coupled from said one to the other of the fibers.
7. An optical fiber coupler as defined in claim 2 in which the
rates of taper of the fibers in the region of said waist are
adiabatic, the minimum diameters and the refractive indices of the
cores and cladding in the region of the waist portion being such
that: V.ltoreq.1 where ##EQU4## in which V is the normalized
frequency (local) of the coupler (dimensionless),
a is the (local) radius of the core,
.lambda. is the wavelength of an optical signal passing through the
coupler and
n.sub.core and n.sub.cladding are the indices of refraction of the
core and cladding respectively,
but that V associated within the waveguide formed by the
air-cladding interface is greater than a value in which the
antisymmetric transmission mode of the cladding waveguide radiates
out of the cladding.
8. An optical fiber coupler as defined in claim 1 in which the
rates of taper are adiabatic, the minimum diameters and the
refractive indices of the cores and cladding in the region of the
waist portion being such that: V.ltoreq.1 where ##EQU5## in which V
is the normalized frequency (local) of the coupler
(dimensionless),
a is the (local) radius of the core,
.lambda. is the wavelength of an optical signal passing through the
coupler and
n.sub.core and n.sub.cladding are the indices of refraction of the
core and cladding respectively,
but that V associated within the waveguide formed by the
air-cladding interface is greater than a value in which the
antisymmetric transmission mode of the cladding waveguide radiates
out of the cladding.
9. A unidirectional optical fiber coupler comprising a pair of
virtually identical fibers each comprised of a core and cladding,
the cladding of one being fused to the cladding of the other at a
narrow waist portion in which the diameters of both the core and
the cladding are narrowed, being tapered adiabatically to the waist
region, the diameters of the refractive indices of the cores and
cladding in the region of the waist portion being such that for the
fiber core waveguide: V.ltoreq.1 where ##EQU6## in which: V is the
normalized frequency (local) of the coupler (dimensionless),
a is the (local) radius of the core,
.lambda. is the wavelength of an optical signal passing through the
coupler and,
n.sub.core and n.sub.cladding are the indices of refraction of the
core and cladding respectively,
but that V associated with the waveguide formed by the air-cladding
interface is greater than a value in which the antisymmetric
transmission mode of the cladding waveguide radiates out of the
cladding.
10. A coupler as defined in claim 9 further including means for
bending the coupler to selectively couple an optical signal passing
along one fiber into the other.
11. An optical fiber coupler as defined in claim 10 including means
for restraining the fibers in either one of a U or S curve in which
the waist is within the curve, the radius or radii of curvature of
the curve being selected whereby an optical signal passing through
one of the fibers is substantially coupled to a predetermined one
or the other of the fibers, and means for varying the radius of the
curve, whereby the optical signal is switched and is substantially
coupled to the other of the fibers.
Description
This invention relates to optical signal transmission and in
particular to an optical coupler for single mode optical signals
having a tunable (variable) coupling ratio, and a method of
fabricating the coupler.
Optical signals are carried via light waveguides, commonly referred
to as optical fibers. Each fiber is typically comprised of a core
and a cladding surrounding the core. The index of refraction of the
core is usually higher than that of the cladding to ensure that
light transmission should occur substantially in the core.
Typically transmission within the fiber occurs in one of two ways:
single mode or multi-mode. Often, in the case of multi-mode fibers,
the index of refraction of the fiber is graded from the core to the
cladding to enhance bandwidth of transmission. In contrast, single
mode fibers are more typically step indexed in refractive index
profile.
Optical signal networks often require couplers in which an incoming
optical signal can be switched to one of a pair of output optical
signal transmission paths. The present invention relates to an
optical coupler for single mode excited fibers. The single mode
coupler is unidirectional and uses biconical taper sections, and
has a coupling ratio which is variable under external control.
The present invention also relates to a method of forming the
coupler, in which the biconical nature of the fibers in the
coupling region is maintained; the method provides a
distortion-free welding of the fibers in the coupler. This
improvement leads directly to superior couplers.
Optical couplers have been known for some time. However the known
biconical taper couplers operate differently from the present
invention resulting in their inability to provide a variable
coupling ratio in a coupler using single mode fibers, which is the
result in the present invention.
U.S. Pat. No. 3,931,518 to Miller describes a coupler involving the
extraction an optical signal to a bulk transparent section and then
to a detector. Light is extracted from a multi-mode fiber to the
bulk transparent section, rather than from a single mode fiber.
U.S. Pat. No. 4,019,051 relates to a bidirectional version of the
Miller invention device. While pressure on the fiber is used to
effect coupling, the pressure controls the amount of the mode
conversion obtained.
U.S. Pat. No. 4,307,933 to Palmer et al relates to a multi-mode
assymmetric fiber coupler.
U.S. Pat. No. 4,336,047 to Pavlopoulos describes the use of
multi-mode fiber biconical taper couplers which appears to be
useful for single mode coupling, but the provision of a variable
coupling ratio as in the present invention is not achieved.
Pavlopoulos describes the use of metal oxide material incorporated
in, on, or about the fibers prior to fusing them. The use of this
material is described for the purpose of slowing the fusion
process. In addition, the material arises the surface tension of
the glass fiber material. In contrast, in one embodiment of the
present invention finely particulated doped quartz soot is used as
a bonding agent, which is compatible with the glass fiber material
and speeds the fusing process, absorbing heat and wetting the
fibers. Thus in this embodiment the present invention has precisely
the opposite effect from Pavlopoulos. The result is a coupler
having lower loss than the Pavlopoulos coupler.
U.S. Pat. No. 4,008,061 to Ramsey describes a form of single mode
fiber coupler, but not a fused biconical form as in the present
invention. The patented structure cannot be tuned in coupling ratio
in the same way as the present invention.
U.S. Pat. No. 4,264,126 to Sheem describes the use of a "bottle
coupler", a single mode fiber coupler which is different form of
coupler from the fused biconical coupler. The Sheem coupler uses
etching to near the fiber core in order to achieve coupling. The
fabrication process and final form are different from the structure
of the present invention.
According to one embodiment of the present invention, two similar
optical fibers are hard fused together. To aid the achievement of
fusing a mixture of high purity ethanol and silica soot can be
applied to a predetermined length of each of the optical fibers.
Heat is applied, whereby the silica soot (if used) wets the fibers,
fusing the cladding together. The fibers are then pulled while the
heat is applied to the fused portion; the fused portion is thereby
narrowed to define a waist. As the fibers are pulled, the diameter
of the cladding and the core at the waist reduce, the diameter of
the core portion reducing to virtually zero. The fibers taper
toward the waist on each side. At a predetermined waist diameter,
tension is stopped and the fibers are allowed to cool.
The fibers are mounted on a base or retainer and bent in a U or
S-shape. The characteristic switching radius of the U or S is
determined by applying light through an input fiber and fixing it
in a position at which slightly varying the radius causes the light
to switch to output either one or the other fiber, but does not
cause an undue increase in the coupler's excess loss. By varying
the bending radius, a variable coupling ratio between the fibers is
achieved.
In accordance with an embodiment of the invention, an optical fiber
coupler is provided, comprising a pair of virtually identical
optical fibers fused together at a narrowed waist region, each
fiber being formed of a core and cladding, each being tapered
toward the waist in such a manner as to ensure adiabatic
propagation of light in the structure. An optical signal carried by
one fiber first passes through a decreasing taper region and then
passes through an increasing taper region of one or the other fiber
(or both). The decreasing taper rate of the input portion of one
fiber is such that the optical signal radiates out of the core
(where V=1 locally) and into the cladding, as it approaches the
waist. The increasing taper rates of the output portions of the two
fibers are such that a predetermined coupling ratio is obtained.
The coupler is bent in the region of the waist whereby a coupling
ratio can be selected between the incoming fiber portions having
the decreasing taper and the outgoing fibers having increasing
tapers.
According to a further embodiment of the invention, an
unidirectional optical fiber coupler is provided comprising a pair
of virtually identical fibers each comprised of a core and
cladding, the cladding of one being fused to the other at a narrow
waist portion in which the diameters of both the core and the
cladding are narrowed. The rates of taper to the waists are
adiabatic, and the diameters of the cores and cladding in the
region of the waist portion are such that: V.ltoreq.1 where
##EQU1## in which
V is the normalized frequency of the coupler (dimensionless),
a is the (local) radius of the core,
.lambda. is the wavelength of an optical signal passing through the
coupler, and
n.sub.core and n.sub.cladding are the indices of refraction of the
core and cladding respectively, but V for the waveguide formed by
the air-cladding interface is greater at the coupler waist than a
value in which the antisymmetric transmission mode of the signal
carried in the cladding radiates out of the cladding.
It is important to note that in contrast to all of the prior art
structures, in the present invention single mode transmission in
one fiber is converted to multi-mode at the waist region where for
the fiber core waveguide V.ltoreq.1, that is, where the optical
signal radiates out of the core into the cladding, to be carried by
the cladding, and can then couple almost entirely into the cladding
of the adjacent fiber depending on the relative indices of
refraction and the radii of the cores and cladding, the taper
angles and the structure length as well as the refractive index of
the medium surrounding the entire structure. In the increasing
taper portion of one or the other fibers the optical signal carried
in the cladding is converted back adiabatically in the taper to
single mode transmission carried by the core of the fiber.
It should be noted that while each of the fibers is to all intents
and purposes identical to the other with identical indices of
refraction, the indices of refraction vary at the point of bend due
to the stresses induced therein. The cladding of one fiber will be
under relative compression while the cladding in the other will be
under relative tension, thus effecting variation in the indices of
refraction.
A better understanding of the invention will be obtained by
reference to the detailed description below in conjunction with the
following drawings, in which:
FIG. 1A illustrates the cross section of a fiber coupler used in
the present invention,
FIG. 1B illustrates an axial cross section of a pair of fibers
prior to fusing,
FIG. 1C illustrates the cross section of a pair of fused fibers at
the waist section,
FIG. 2 is a schematic representation of a single mode fiber coupler
with a tunable coupling ratio in accordance with the present
invention,
FIG. 3 is a graph of fiber diameter against the length of the
coupler, and
FIG. 4 is a graph of optical power realized in each output portion
of the fibers of the coupler against axial displacement as the
coupler is bent.
Turning first to FIG. 1A, an elongated cross section of the fiber
coupler according to the present invention is shown. A pair of
fibers 1A and 1B are each comprised of a core 2A and 2B
respectively each covered by a cladding 3A and 3B respectively. A
waist portion, generally shown by portion W, is fused together
gently. This is effected for ease of fabrication by applying a
mixture of silica soot in high purity ethanol to the surfaces of
the fibers, holding them together, and heating the waist section.
Silica soot can be made by chemical vapour deposition (CVD), a well
known process. It is preferred that the silica soot should be doped
with boron, in order that its melting temperature should be lowered
below that of the silicon optical fibers. In addition, the index of
refraction of the soot should be less than that of the cladding in
order to retain the optical signal within the cladding.
The fibers are then heated and pulled to form a biconical taper
coupler. An optical signal is applied, and pulling is continued
until overcoupling is obtained. Referring to FIG. 1 and expansion
view 9 of FIG. 2, assuming that an optical signal is applied to one
fiber and arrives in the fiber from the left, it must pass through
a decreasing taper region I followed by an increasing taper region
O in one or both of output portions of the fibers to the right.
FIG. 1B illustrates the two identical fibers 1A and 1B prior to
fusing and narrowing at the waist, and being comprised of cores 2A
and 2B and cladding 3A and 3B.
FIG. 1C illustrates a cross section of the waist portion of the
fused fibers in which several significant aspects of the invention
can be observed. The diameters of the claddings 3A and 3B have been
reduced from the diameters in FIG. 1B. The diameters of the cores
2A and 2B have been reduced to virtually nil (indeed, they can be
reduced to nil). Further, the fibers have been hard fused, that is,
they have retained substantially their original profile. Soft
fusing would have changed the overall profile to the structure
similar to an ellipse or dumb-bell shape, while in the present
invention the entire circular profile of each of the fibers has
been retained. The biconical nature of the coupler is thus
preserved in the present invention.
The coupler is cooled and is bent into a broad U-shape or S-shape
until minimum or zero coupling is obtained from the first to the
second fiber. In some cases the S-shaped bend may be out of
alignment with the plane of the fibers. The bent coupler is mounted
(either prior to or after bending), and by straightening the
coupler the coupling ratio can be varied.
It should be noted that in case depressed cladding fibers are used
to form the couplers, the tapered and waist portions of the fibers
should be etched to remove the outer cladding in the area of the
fiber strand where fusion will take place, prior to the step of
applying the silicon soot.
FIG. 2 illustrates an embodiment which demonstrates the invention.
A He-Ne laser 4 provides light at its characteristic wavelength
into one end of fiber 1A. A coupler 14 is fabricated as described
earlier with fiber 1B, the bent portion (also seen in expansion
view 9) being mounted on a flat spring base 4. The end of fiber 1A
emits a light spot 5 on a screen 6. A micrometer 7 or equivalent
apparatus is used to bend the spring 4, thus straightening or
otherwise stressing the bend in the coupler. The light is coupled
in the coupler from fiber 1A to fiber 1B, thus dimming light spot 5
and illuminating light spot 8. The degree of energy coupled between
fibers 1A and 1B can in this way be dynamically varied by
micrometer 7.
Micrometer 7 can of course be replaced by a solenoid, a
piezo-electric crystal, or any other controllable apparatus which
can provide a bending stress or a change in refractive index at the
waist of the coupler.
While on the surface there may appear to be many similarities to
prior art structures, there are highly significant aspects which
cause the present invention to differ substantially in structure,
operation and result from that of the prior art.
It was pointed out earlier that the present invention is directed
to optical waveguides in single mode excitation. It should be noted
that the decreasing taper portion of the coupler I causes the
signal input fiber core to radiate into its cladding. This is
believed to excite a combination of the two lowest order modes of
the entire fused-cladding waveguide structure. The dephasing of
these two modes in effect transfers power from the input fiber
cladding region of the waveguide structure to the adjacent fiber
cladding region in the coupler. The coupling ratio is determined by
the phase difference between the odd and even modes during
recapture in the increasing taper region of the coupler.
The present structure and mode of operation thus differs
significantly from those couplers of the prior art, in which
coupling is achieved through evanescent wave coupling between
cores.
The decreasing taper rate of the input fiber must be such that the
optical signal radiates out of the core and into the cladding with
minimal loss of light from the structure. The increasing taper rate
of the output portions of both fibers must be such that a
predetermined coupling ratio is obtained. In the increasing taper
portion the optical signal is converted back to single mode, and is
carried by the core of the output portion of the selected fiber, or
of both fibers.
The rates of taper to the waists must be adiabatic that is, the
spot size must grow gradually within the fiber, but must stay
within the cladding.
For the above effect to be achieved, it has been determined that
the normalized frequency V for the core-guided light must be equal
to or smaller than unity in the structure, i.e. V.ltoreq.1 where
##EQU2## where
V is the normalized cut-off frequency of the coupler (and is
dimensionless),
a is the (local) radius of the core,
.lambda. is the wavelength of an optical signal passing through the
coupler, and
n.sub.core and n.sub.cladding are the indices of refraction of the
core and cladding respectively.
It should also be noted, that V for the waveguide formed by the
air-cladding interface should be greater at the coupler waist than
a value in which the antisymmetric transmission mode of the
cladding waveguide radiates out of the cladding.
The result is radiation of the symmetric mode out of the core and
into the cladding but the antisymmetric cladding mode remains
bound. In the region of the coupler in which the symmetric mode and
antisymmetric modes of the cladding propagate, the ambient (air or
coating) surrounding the cladding forms an external cladding to the
fiber. In the region of the waist, where the signal is carried in
the cladding, the core can disappear as it plays no part in the
transmission of the signal.
The above limitations ensure that coupling is effected from
cladding to cladding, and that there is no evanescent coupling
between core and core as in the prior art couplers.
Various typical wavelengths carried by the fiber used to make the
coupler are e.g. 0.85 microns, 1.3 microns or 1.5 microns, in the
HE.sub.11 mode. Typical fiber dimensions are core radius of about 3
to 10 microns and cladding radius of about 75 to 150 microns (prior
to etching). Typical indices of refraction are 1.451 for the core
and 1.44 for the cladding.
FIG. 3 illustrates a fiber profile in the region of the coupler in
a successful prototype. The normalized cut-off frequency V (defined
as V=1) of the core, where the transition from core transmission to
cladding transmission or vice versa, may be observed over a length
of about 33/4 millimeters (the length demarcated by the positions
X--X along the abscissa).
FIG. 4 is a graph illustrating the measured optical power carried
by a pair of fibers of a laboratory prototype of a coupler in
accordance with the present invention as the coupler is bent (or
the radius of curvature of a bent coupler is straightened). Curve A
illustrates the optical power carried by one fiber while curve B
illustrates the optical power carried by the second. Curve T
illustrates the total power, which, it may be seen, is nearly a
constant. The shapes of the two curves illustrate that one is
virtually the reciprocal of the other. It may also be seen that
over a substantial portion of the curve a variable coupling ratio
is achieved.
Clearly the present invention constitutes a substantial advance in
the art, providing a single mode fiber biconical taper coupler
which has a controllable, variable coupling ratio, and which does
not use core to core coupling as in prior art couplers.
A person skilled in the art understanding this invention may now
conceive of variations in design using the principles described
herein. All are considered to be within the sphere and scope of
this invention as defined in the claims appended hereto.
* * * * *